Volatile Components of Roasted Peanuts. The Major Monocarbonyls

graph equipped with a flame ionization detector. ... each major peak and at the apexes of the very minor ... Figure 1 shows a typical gas chromatogram...
0 downloads 0 Views 569KB Size
PEANUT COMPONENTS

Volatile Components of Roasted Peanuts. The Major Monocarbonyls and Some Noncarbonyl Components M. E. MASON,' BOBBY JOHNSON,' AND MYNARD

Condensates obtained from roasted Spanish peanuts using high vacuum distillation and cryogenic trapping procedures possessed typical roasted peanut aroma. Analyses using gas chromatographic and qualitative chemical techniques revealed the presence of several aldehydes. Analysis of the total condensates and carbonyl compounds regenerated from their 2,4-dinitrophenylhydrazones using a combination gas chromatograph-mass spectrometer (GC-MS) and

The presence of carbonyl as well as other classes of compounds in roasted peanuts was revealed by Pickett and Holley (7). Recently, the presence of mono- and dicarbonyl compounds has been confirmed (IO). Pattee, Beasley, and Singleton (6) found carbonyl compounds as well as ethanol and ethyl acetate to be volatile constituents of peanuts which had been treated to produce off-flavors. However, little has been done to identify the specific compounds associated with the typical flavor from roasted peanuts. The authors positively identified several of the volatile carbonyl compounds found in roasted Spanish peanuts. These identifications should serve to define the typical roasted flavor chemically. Brief comments are made concerning the possible contributions of some of the components to the total flavor. Reagents and Materials

2,4-Dinitrophenylhydrazine, reagent grade, Eastman Kodak Co., Rochester, N. Y . Gas-Chrom Q, 100 to 120 mesh, Applied Science Laboratories, Inc., P. 0. Box 140, State College, Pa. Carbowax 20M, Apiezon L, and Chromosorb W, 60 to 80 mesh, Varian Aerograph, 2700 Mitchell Drive, Walnut Creek, Calif. Glycerol, U. S. P., heavy-grade, Pro cter & Gamble Co., Cincinnati, Ohio. Phenylacetaldehyde, benzaldehyde, acetaldehyde, isovaleraldehyde, 3-methylbutanal, K and K Laboratories, Plainview, N. Y . Methylene chloride, nD20 1.4238, Aldrich Chemical Co., Milwaukee 10, Wis., redistilled at 40" C. Department of Biochemistry, Agricultural Experiment Station, Oklahoma State University, Stillwater, Okla. * Continental Oil Co., Ponca City, Okla. 66

J. AGR. FOOD CHEM.

c. H A M M I N G ~

thin-layer chromatographic analysis of the 2,4D N P H derivatives resulted in the positive identification of acetaldehyde, isobutyraldehyde, benzaldehyde, and phenylacetaldehyde, and tentative identification of 2- and 3-methylbutanal and 3methyl-2-butanone. Ethyl acetate, toluene, and N,N-dimethylformamide were tentatively identified but their presence in the carbonyl regenerates remains unexplained.

Skellysolve B., Skelly Oil Co., P. 0. Box 36, Kansas City, Mo. Ethyl acetate, reagent grade, J. T. Baker Chemical Co., Phillipsburg, N. J. Benzene, reagent grade, Fisher Scientific Co., Fair Lawn, N. J. a-Ketoglutaric acid, Calbiochem, 3625 Medford St., Los Angeles, Calif. Procedures

Preparation of Condensates from Roasted Peanuts. This procedure involved collecting volatile compounds, at pressures of about mm. of Hg, from roasted Spanish peanuts ground to a fine slurry in glycerol and water. The glycerol was found free of carbonyl compounds by testing with the 2,4-dinitrophenylhydrazin,reagent. Liquid nitrogen was used as the coolant. I n order to eliminate water for preparative gas chromatography, the aqueous condensates were extracted three times with 1/10 volumes of methylene chloride. About 3 ml. of condensate were obtained for each pound of peanuts extracted. The extracts were combined and. reduced to about one tenth their former volume on a rotary evaporator, and 200-pl. quantities separated by gas chromatography. In this procedure, the solvent peak made collection of the first five or more components impossible. However, the removal of the water which was required for chromatography on the combination gas chromatography-mass spectrometer (GCMS) instrument was achieved. The aqueous condensates were chromatographed without previous extraction (Figure 1) using the Perkin-Elmer 800 gas chromatograph equipped with a flame ionization detector. Preparation of 2,4-Dinitrophenylhydrazone (2,4DNPH) Derivatives. T o the thawed, aqueous condensates freshly removed from the liquid nitrogen traps were added 10 volumes of a saturated solution of 2,4dinitrophenylhydrazine in 2 N HCI. The precipitate

was collected by filtration and dried over anhydrous calcium sulfate in a vacuum desiccator. Regeneration of Carbonyls from Their 2,4-DNPH Derivatives. Regenerations were performed according to the procedure of Ralls (8) with some modification of regeneration tubes ( 4 ) . For gas chromatography, 6 t o 40 mg. of the mixture of the 2,4-DNPH derivatives were weighed and triturared with three times their weight of a-ketoglutaric acid ;and the mixtures transferred to the regeneration tubes. Using a gas-tight syringe, carbonyl vapors were taken from the hot tube immediately after regeneration and injected into the gas chromatographmass spectrometer (GC-MS instrument). Gas Chromatography. Relative retention times were obtained b,. gas chromatography of aliquots of the aqueous condensates on a Perkin-Elmer Model 800 gas chromatograph equipped with a dual hydrogen flame detector using a 20-foot X l/S-inch 0.d. (0.062-inch i.d.) stainless steel column packed with 2 0 z w./w. Apiezon L o n Chromosorb Mi, 60 to 80 mesh. Nitrogen was the carrier gas and flow rates of 35 to 40 ml. per minute were used; column temperatures were programmed at 6 " C. per minute from 100" to 200" C . Retention times ofthe unknowns were normalized by setting that of butanal (internal standard) equal to 1.00. Preparative separations of unknowns 18 and 21 (Figure 1) were made on a 20-foot x 1/4-inch (0.25-inch o.d., 0.16-inch i.d.) diameter column packed with the same material as the I/S-inch column (0.12-inch o.d., 0.062-inch i.d.). An F and M Model 500 gas chromatograph provided with a four-filament hot-wire detector was the instrument used. For collection of eluted compounds, glass U-tu'cqes were constructed so that the inlets could be inserted directly into the exit port of the thermal detector through a silicone rubber septum. The construction of these traps has been described (5). Gas Chromatography-Mass Spectrometry. The combin2tion gas chromatograph-mass spectrometer (GCMS) instrument used was a prototype of the LKB 9000 mass spectrometer constructed in the laboratories of Ragnar Rq hage, Karolinska Institutet, Stockholm, Sweden. To obtain the gas chromatographic separations for mass spectral analyses, gas from regenerated 2,4-DNPH derivatives was introduced onto a 0.25-inch 0.d. X 24-foot glass column packed with 5 % w./w. Carbowax 20M on Gas-Chrom Q. Helium at a flow of 35 ml. per minute was used and temperatures used are specified elsewhere. Mass spectra were taken at successive points along each major peak and at the apexes of the very minor peaks. In this way, the homogeneity or heterogeneity of peaks representing major components could be confirmed. Thin-Layer Chromatography of 2,4-DNPH Derivatives. Procedures used for analytical and preparative thin-layer chromatography were those outlined by Dhont and D e ROOY(2) and Ruffini (9). Solvent systems used were as follows: solvent system I, 5 ethyl acetate in benzene (v./v.); solvent system IT, ethyl acetate and Skellysolve B, 1 :2 (v./v.); solvent system 111, benzene and Skellysolve B, 3:l (viv.). Visible spectral analyses o n preparatively separated derivatives

z

were performed o n the Cary Model 14 spectrophotometer in the chloroform solvent in which they were eluted. Results and Discussion Figure 1 shows a typical gas chromatogram of the aqueous suspension (condensate) removed from roasted peanuts by vacuum distillation in the manner described above. No quantitative significance was attributed to this chromatogram because of the heterogeneity of the sample. When this suspension was extracted with methylene chloride and the extract was chromatographed on the preparative column using the hot-wire detector, several carbonyl compounds were detected among the eluting components by bubbling the effluent gas through a saturated solution of 2,4-dinitrophenylhydrazine in 2 N HC1. These included components labeled 1 to 5 , 18, and 21. Components 1 to 5 possessed odors similar to aliphatic aldehydes while component 18 had a cinnamon-like odor. Component 21 had an odor reminiscent of oil of roses. Subsequently, relative retentions (retention of 1butanal = 1.00) of some of the c o m m x aliphatic aldehydes and of benzaldehyde (cinnamon-like) and phenylacetaldehyde (oil of roses) were determined. Component 1 corresponded to acetaldehyde which had an average relative retention time, RR1, of 0.50, while component 3 corresponded to isobutyraldehyde which had a n RR1 of 0.86; component 4 corresponded to 3-methylbutanal-RRt of 1.37. Even though authentic 2-methylbutanal was n3t available, the compound was made in situ from a heated mixture of isoleucine and ninhydrin in citrate buffer, and a sample of the vapor was gas chromatographed using headspace sampling techniques. This aldehyde had an R R t (1.41) slightly higher than that of 3-methylbutanal but the same as that of component 5 . Authentic benzaldehyde had an average RRl of 5.56, while that of phenylacetaldehyde was 6.40. These corresponded to the average R R l of components 18 and 20 (Figure 1). Relative retention values (R, componentlRf 3-pentanone) from thin-layer chromatograms and wavelengths of maximum absorption (A,, ) for the 2,4-DNPH derivatives of isolated components and authentic standards are listed in Table I. , , ,A data agreed well with published values (3). The data confirmed the presence of acetaldehyde, phenylacetaldehyde, and one or both of the isomeric methylbutanals. Isobutyraldehyde and a ketone (Amny 363 mp) migrated together in all solvents and the presence of the ketone was deduced from the A,,, of this mixture. However, a shoulder at about 356 m p indicated the presence of the aldehyde in a somewhat lesser amount. The conclusion that the ketone was 3-methyl-2butanone was based on the agreement of the A,, anJ relative retention of the unknown thin-layer spot with that of the derivative of the authentic compound (Table I). Also, 3-methyl-2-butanone had almost exactly the same gas chromatographic RRl as 3-methylbutanal and would have eluted under the same peak as the isomeric niethylbutanals (Figure 2). VOL. 15, NO. 1, JAN.-FEB. 1967 67

21

I

4

II

24

rl

30

t

J

IO

0

I \I\

77

I

l

l

,

I

,

,

,

Ulc ,

,

,

,

,

,

,

,

,

,

4 0 38 36 34 32 30 28 26 24 22 20 18 16 14 I2 IO 8 6

(

,

4

2 0

MINUTES

Figure 1. Gas chromatogram of 1 pl. of aqueous condensate from roasted Spanish peanuts

Table 1.ERelative Retentions ( R f Component/Rf $Pentanone) and Wavelengths of Maximum Absorption by Thin-Layer Chromatography of Isolated Components and Authentic Reference Standards Ln,, Solvent I Solvent I1 Solvent 111 2,4-DNPH ComComComComDerivatives ponent Standard ponent Standard ponent Standard ponent Standard Acetaldehyde 355 354 0.81 0.81 0.75 0.75 0.82 0.82 Isobutyraldehyde and 363 with 357 1 .oo 1 .oo 1.01 3-Methyl-2-butanone shoulder 363 1 .oo 1.02 1.01 1.01 1.03 1 .oo a t 356 3-Methylbutanal 358 358 1.06 1.03 1.08 1.06 1.01 1.02 Phenylacetaldehy de 355 355 0.99 1 .oo 0.99 1 .oo 0.83 0.80

20

15

IO

5

0

TIME -MINUTES

Figure 2. Gas chromatograph-mass spectrometer (GC-MS) strip chart obtained by chromatography of 250 p1. of vapors regenerated from 9 mg. of 2,4-dinitrophenylhydrazone derivatives

From the relative amounts of 2,4-DNPH derivatives present after separation by thin-layer chromatography, the isomeric methylbutanals and mixture of isobutyr68 J. AGR. FOOD CHEM.

aldehyde and 3-methyl-2-butanone were estimated to be present in approximately equal amounts and collectively constituted most of the monocarbonyls. However, the amount of acetaldehyde derivative present was much larger than was predicted from the GCMS analysis of the regenerated carbonyls. The amount of phenylacetaldehyde derivative was small and very small amounts of other monocarbonyl derivatives occasionally visible on the thin-layer chromatograms were probably due to benzaldehyde and other monocarbonyl compounds whose presence in minute amounts was indicated by mass spectral evidence. Biscarbonyl derivatives possessed very low retentions in the solvents used and their identifications will be the subject of another paper. In order to gas chromatograph the monocarbonyl compounds free of biscarbonyls and other classes of compounds such as the pyrazines (3,they were regenerated from the mixture of 2,4-DNPH derivatives as described. Figure 2 shows a chromatogram from the GC-MS instrument obtained by injecting 250 pl. of vapors from 9 mg. of regenerated 2,4-DNPH derivatives; the column was operated at 70" C. Only three peaks other than air were obtained at the attenuations used and one of these (component 2) was extremely small. Thus, components 1 and 3 were by far the major monocarbonyl compounds present as determined by this procedure. Table I1 shows complete spectra of these

two major components, along with reference standard spectra taken at or near the apexes of the peaks; intensities less than 1% of the major fragment were omitted. Intensities of major ions in successive spectra taken of the two major peaks were included, also. The successive spectra were obtained at points marked 1, 2, 3, etc., which appear as discontinuities on the curves in Figure 2. Successive spectra showed that component 1 was homogeneous and the spectrum at the apex corresponded to that of isobutyraldehyde confirming evidence obtained from thin-layer chromatography. Conversely, successive spectra showed that component 3 was heterogeneous, consisting of 2-methylbutanal but containing some other carbonyl compound (probably 3-methylbutanal) under the trailing edge of the peak. This was manifestfed by sharp increases in relative intensitites of m/e values of 27, 41, 44, and 58 in spectra

4 and 5 of component 3 while other fragment intensities were nearly the same as in the first three spectra. By consulting the reference standard spectra for 2-methyland 3-methylbutanal shown in Table 11, the major differences in intensities of fragments, relative to an mje of 57, were at m/e values of 27, 44, and 58. Thus, the increase of relative amounts of these fragments in spectra 4 and 5, especially m/e = 44, was probably due to the presence of some of the 3-methyl derivative. In a subsequent chromatogram, discussed later in the text, component 2 was found to be ethyl acetate. The reference standards shown in Table 11, with the exception of 2-methylbutanal, were obtained by analyzing commercially available compounds on the GC-MS instrument. That of 2-methylbutanal was from Dow uncertified spectra ( I ) . In order to obtain good spectra of minor monocarbonyls present, 40 mg. of the 2,4-DNPH derivatives

Table 11. Mass Spectra of Components from GLC Separation of 250 pl. of Vapors from Regenerate of 9-Mg. 2,4-DNPH Derivatives Relative Intensities -~ Significant Ions of Successive Spectra Complete Spectra of Components and.Standards ~ ~ _ _ _ ~ Spectra of Component 3 -Isobuf yr2-Methyl- 3-MetIg I- Spectra of Cornpone& 1 2 3 1 2 3 4 5 mle 1 aldehyde 3 butanal butanal 15 1 5 4 4 1 5 1.5 5 4 6 0 :9 3 0 26 3 3 3.9 29 2 3 9 . 6 3 8 . 8 39.5 2 3 . 4 2 9 . 2 31.5 37.3 33.9 48 5 26.9 35 5 27 39 6 13 0 1 1 28 11 3 19.5 8.9 26 2 29 25 8 91 2 25.8 2 4 . 2 2 4 . 6 8 6 . 5 91.2 8 8 . 8 97.1 95.5 22 7 88 2 30 1 1 3.2 3 6 31 1 0 1.5 1 8 1 5 2 6 37 2 1 4 4 5 5 38 3 3 1.9 28 5 17.0 39 18 9 20 5 18.9 19.8 19.2 18 9 40 4 1 2 5 5 8 2.2 3 3 81 . O 87 2 65 0 41 64 3 6 4 . 3 6 4 . 0 6 7 . 5 8 1 . 5 8 7 . 2 8 9 . 8 107.0 125.0 79 0 5 7 42 8 3 9 4 4.3 16 8 10.0 43 100 0 100 0 15 7 73 5 1 0 0 . 0 100.0 1 0 0 . 0 1 5 . 9 1 5 . 7 2 0 . 5 3 6 . 5 54.5 7 5 44 4 6 4.5 7 . 5 17.0 42.2 67.5 8 0 100 0 1.3 15 5 45 0 5 2 2 1 8 3.2 50 1 6 2 4 1.5 51 2 3 1 8 1 6 53 10 1 4 3 9 3.0 5 2 2 4 55 1 5 8 9 8.5 5 4 56 3 0 8 3 8.8 57 2 9 23 5 4 9 100 0 1 0 0 . 0 2.9 2.9 2 . 6 100.0 100.0 100.0 100.0 100.0 58 10 51 8 4 6 65.0 7 1 . 5 7 2 . 5 7 5 . 5 8 5 . 4 94.0 72 5 59 2.6 3 3 3 5 61 1 6 67 1 6 68 2 0 69 2 2 70 1 4 71 1 0 1.2 6.4 4.5 21 2 72 52 8 42.6 1 9 52.8 49.2 48.8 73 2 6 5.8 4 4 74 4.8 4.3 85 2 5 86 11.1 15.3 6 7 11.0 11.1 10.8 12.2 13.4 87 0 8 0.9 3.1

VOL. 15, NO, 1 , JAN.-FEB. 1967 69

w

v)

z

0 0 v)

a Y

I

50

60

40 TIME

- MINUTES

20

Figure 3. Gas chromatograph-mass spectrometer (GC-MS) strip chart obtained by chromatography of 5 ml. of vapors regenerated from 40 mg. of 2,4-dinitrophenylhydrazonederivatives

130

Componenf i Acetaldehyde

9080 -

70 60 -

ae :

50-

,

c W e

-c w

40-

-0

aw:

30 -

O

20

Q

1

1

40 mle

50

A 20

1

4

7

50

mle

Figure 4. Mass spectra of unknown 1 (Figure 3) and acetaldehyde

were regenerated and 5 ml. of the vapors introduced onto the column attached to the GC-MS instrument. This represented about 100 times the amount of material used for the chromatogram shown in Figure 2. Figure 3 is the chromatogram obtained showing the number of 70 J . AGR. FOOD CHEM.

components separated (arabic numerals) and the number of spectra taken o n each peak (discontinuities). Again discontinuities on the curves, except for one point where an attenuation was made, marked the points at which successive spectra were taken. The temperature of the column was initially 50" C., but was raised abruptly at several points during the chromatogram as shown in the figure. Components 2 and 4 of Figure 3 were found to be the same as 1 and 3 of Figure 2 (isobutyraldehyde and 2methylbutanal mixed with 3-methylbutanal and 3methyl-2-pentanone). Spectra of components 1, 3, 7, and 10 compared well with spectra of reference standards (Figures 4, 5 , 6, and 7) obtained under similar conditions, except for that of toluene which was taken from API spectra. Component 1 proved to be acetaldehyde, which confirmed evidence obtained from thin-layer chromatography and visible spectral analysis. Obviously, from the size of peaks in Figure 3, all monocarbonyls except isobutyraldehyde and the isomeric methylbutanals are minor constituents of the condensates, confirming conclusions drawn from thinlayer separations. Just how such compounds as ethyl acetate, toluene, and dimethylformamide could appear in a regeneration of carbonyl compounds was not immediately obvious, Since water was the only solvent used in making these derivatives, the most likely explanation was that these compounds were occluded in or adsorbed to the 2,4-DNPH derivatives in such a manner that they were not freed during the washing process. This explanation seems likely because ethyl acetate has been reported as a constituent of volatiles from roasted peanuts by Pattee, Beasley, and Singleton (6). Other possibilities are that such compounds arose during the regeneration process or were impurities in the 2,4-DNPH reagent or a-ketoglutaric acid used in the regeneration; these seem less likely, however.

100..c

90 -

80Ethyl

Component 3

Acetate

713 60 -

s ; ,

--

53-

a2 c E

-'1 4 0 -

.-

-

0 a2

U

30 20 -

20

30

50 60 m le

40

Figure 5.

70

80

90

20

30

40

50

60

70

80

90

ml e

Mass spectra of unknown 3 (Figure 3) and ethyl acetate

liso-

30 -

30 70 Toluene

60-

ae 50' E

c

e

.-%!

40-

-

c 0

d

30 -

20-

IO

-

0 1 20 30

L 50

60

m /e

70

80

90

100

20

30

40

50

60

70

80

90

-

100

ml e

Figure 6. Mass spectra of unknown 7 (Figure 3) and toluene VOI.. 15, NO. 1, JAN.FEB. 1967 71

100N , N - Dirnethylforrnomide

component 10

9080 -

70 60 8 r

.z 50-

a2

-c w

-w

40-

0

a

30-

50 m le

rnfe

60

70

60

Figure 7. Mass spectra of unknown 10 (Figure 3) and N,N-dimethylformamide

Table 111. Tentative Identities of Minor Components Shown in Figure 3 Based on Mass Spectral Data Component Tentative Identity 5 3-methyl fur an 6 Pentanal Thiacyclohexane 8 9 2-Methylbutanal 11 3-Methylthiacyclohexane 12 4-Meth ylthiacyclohexane 13 Heptanal 14-1 8 Unknown

Some of the minor constituents may be of major importance for flavor-for example, the authors believe that phenylacetaldehyde will prove of considerable importance to the sweet bouquet of roasted peanut aroma. Table I11 contains a listing of remaining components in Figure 3 for which tentative identities were obtained from the mass spectra. Background was relatively high in spectra of the latter five components (14 to 18 in Figure 3) and probable identities were not discernible. Benzaldehyde and phenylacetaldehyde were not present in the regenerated vapors in sufficient amounts to obtain good mass spectra from the Chromatographic 72 J. AGR, FOOD CHEM

peaks. Consequently, methylene chloride extracts of the condensate were chromatographed on the GC-MS instrument. Mass spectra obtained from components corresponding to peaks 18 and 21 of Figure 1 are listed numerically in Table IV along with reference standards obtained from commercially available chemicals. These data confirmed thin-layer and visible spectral evidence for the presence of phenylacetaldehyde. Although objective determination of the contribution of these monocarbonyl compounds to typical roasted peanut flavor has not been possible yet, their odors and abundance suggest they must exert a very important contributory effect to over-all flavor. For instance, when phenylacetaldehyde was removed from the total condensate by gas chromatography, and the remaining components were recombined, the sweet background or bouquet of roasted peanut aroma was lost. When the low molecular weight aldehydes were removed, the harsh aroma usually associated with warm freshly roasted peanuts was lost. In both cases, dramatic differences in intensity and type of aroma resulted. Since these were removed by gas chromatographic separation, possibly small amounts of other components were lost also which could have an effect on the aroma. However, compounds having typical roasted and nutty aromas were still present and the roasted aromas have been attributed to certain pyrazine compounds and their mixtures (5).

-

Table IV. Mass Spectra of Compounds Corresponding to Components 18 to 21 in Figure 1and Reference Standards Relative Intensities BenzComPhenylComaldehyde ponent acetaldehyde ponent standard 18 standard 21 mle 29 37 38 39 41 49 50 51 52 53 61 62 63 64 73 74 75 76 77 78 89 90 91 92 93 105 106 107 120 121

3.8 4.0

11.1 3.8

9.0

15.2

4.1 23.5 44.8 15.8 2.1 2.0 2.5 3.9

5.7 28.1 42.2 12.0 6.1 3.1 2.3 3.2

5.1 13.1 6.7 7.5 100.0 26.8

25.3 15.6 7.5 100.0 19.0

1.2 7.3 2.1

1.9 10.3 5.4

2.5 4.5

3.4 6.6

A ckno M? ledgment

Purchase and maintenance of the GC-MS instrument were made possible through National Science Foundation Grant G B 3482. The authors are indebted t o Ralph Matlock for furnishing peanuts of known grade and genetic background and t o Rolf Hjalm for technical assistance with the mass spectrometer.

Literature Cited

2.4 5.8 1.3

2.5 6.0 2.2

(1) ASTM Committee E-14 File of Uncertified Mass Spectra, A. H. Struck, Chairman, Perkin-Elmer C o n . Norwalk. Conn. (Spectra data from Dow .~ C h i m k a l Co.). (2) Dhont, J. H., De Rooy, C., Analyst 86, 74 (1961). (31 Jones. L. A.. Holmes, J. C . . Selianan. R. B., Anal. Chem. 28, 191 (1956). ’ (4) Mason, M. E., Johnson, Bobby, Hamming, M. C., Ibid.. 37. 760 (1965). (5) Mason: M. E., Johnson, Bobby, Hamming, M. C., J. AGR.FOOD CHEM.14,454 (1966). (6) Pattee, H. E., Beasley, E. O., Singleton, J. A., J . Food Sei. 30, 388 (1965). (7) Pickett, T. A., Holley, K. T., Ga. Expt. Sta. Tech. Bull. 1,1952. ( 8 ) Ralls, J. W., Anal. Cliern. 32, 332 (1960). (9) Ruffini, G., J . Chromatog. 17, 483 (1965). (10) Young, C. T., Holley, K. T., Ga. Expt. Sta. Tech. Bull. N.S. 41, 1965. .

3.6 2.8 100.0 23.8 1.8 96.5 96.8 12.7

The major monocarbonyls probably arose by Strecker degradation of corresponding free amino acids. Information which indicates this to be the case will be the subject of a future paper.

4.7 2.8 100.0 24.0 3.2

109.9 109.9 12.7 26.5 2.6

27.0 4.2

I

Receiced j ) r reciew M a y 19, 1966. Accepted October 21, 1966. Dicision of Agriculturul and Food Chemistry, 152nd Meeting, A C S , New York, N . Y.. September 1966. Work supported by the Oklahoma Agricultural Experiment Station and Corn Products Institute oj Nutrition.

VOL. 15, NO. 1, JAN.-FEB. 1967 73